TRIAGE proposes a training-free visual token acceleration method that decides whether to reuse cached KV or prune based on a token drift threshold, and fixes the recomputation budget to directly control prefill latency.
The paper proposes a structure-preserving quantum representation that maps rigid-body transformations and serial-manipulator kinematics to parameterized quantum circuits, encoding spatial directions in single-qubit states and reconstructing end-effector poses from Pauli expectation values. Using the parameter-shift rule, the manipulator geometric Jacobian can be reconstructed from derivatives of quantum observables, analytically equivalent to the classical result under ideal readout. On this basis, the authors propose a hybrid closed-loop method, Quantum Closed Loop Inverse Kinematics, which uses the quantum-derived pose and Jacobian information to solve inverse kinematics.
The authors propose a keyframe management method for dense maps on memory-constrained robots that directly limits the memory occupied by the resident map and admits and releases keyframes by semantic spatial units such as building rooms and corridors. Across every run on two NVIDIA Jetson boards, the memory budget was never exceeded and model throughput loss was at most 0.8%, whereas without the method the operating system terminates the model or the map. Under that budget, admission by semantic zone localized 6-9 percentage points more query frames than geometric partitioning by the number of matched regions, roughly twice that of RTAB-Map retrieval, with lower ATE.
TinyFlow is a 1.26M-parameter warp-free optical flow network that replaces flow-dependent feature resampling with fixed-window correlation over constant offsets, allowing the entire network to compile as a single static INT8 graph.
AgiBot has released Genie Studio 2.0, an embodied intelligence application platform covering development, deployment, and operations. It consists of three modules—Flywheel, Agent, and Monitor—that connect data collection, model training, evaluation, inference, task orchestration, and robot fleet management.
Why it matters: The original text provides the platform's three major modules and validation data from multiple production lines, which can be used to assess the engineering chain of embodied applications from development to operations.
The ZRCar team demonstrated a custom PCB test on its LEGO Technic-compatible frame: the RGB strip runs a "chase" lighting effect, the headlights turn on, and steering is driven by LEGO components. The video does not provide a side-by-side view of browser-based remote control; the team says another video documents that feature.
This article presents the complete process of deploying ROS LiDAR mapping on an RK3588 edge computing box, covering hardware wiring, system configuration, driver debugging, indoor data collection, and map saving. It also shares troubleshooting methods for common issues, aimed at helping beginners get started.
Based on its "one brain, multiple forms" architecture, Youibot has launched MOMABase, an industrial-grade embodied intelligence base platform that integrates underlying capabilities such as power supply, communication, computing, and interfaces. It can be paired with actuators and sensors and connect to large models such as VLA, shortening the integration and deployment cycle for multi-form robots.
Beijing Humanoid launched the "Huisi Kaiwu App Store" in October. Robot skills are packaged as apps that can be searched, downloaded, and deployed with one click to run on full-size general-purpose robots and other bodies. The store also offers datasets and simulation assets, and the official skill app library is open for free; developers who list skills and have them called for a fee can share in the revenue. Users can search for "Jushen Tiangong" in mobile app markets and enter the app store as developers to use the relevant features.
Afferent-Omni is an open-source visuo-tactile command and control (C2) framework for embodied AI and dexterous manipulation, offering a FastAPI backend, 9 modular routers, 15 REST endpoints, and a 50 Hz closed-loop WebSocket stream.
Comau announced the integration of Nvidia technology into its product portfolio and design processes, including industrial robot controllers using Nvidia Jetson edge computing, connecting its robot product line to Nvidia Isaac Sim, and using the Nvidia software stack to optimize the vision capabilities and control strategies of its intralogistics products.
Qualcomm Technologies has agreed to acquire robotics software company PickNik, which has long maintained the open-source manipulation framework MoveIt. Qualcomm said MoveIt will retain its existing open-source license, community roadmap, and support for third-party hardware, and it plans to integrate MoveIt and the commercial MoveIt Pro more closely with the Dragonwing robotics platform and Arduino VENTUNO Q boards, connecting vision-language-action models with motion planning, manipulation, and real-time control.
The author released a Python SDK that can connect any LLM or agent to an SO-101 or dual-arm YAM robotic arm; in the demo, Gemini 3.7 performs a "stack the yellow blocks" task (video sped up 10x).
All Orbbec Gemini 330 series active stereo cameras perform depth computation and depth-color alignment in-camera on the proprietary MX6800 depth engine ASIC, without using host computing power. Since launching the Gemini 335 and 335L in April 2024, the series has continued to expand, covering different baselines, optical filters, interfaces, and protection ratings.
Hesai has launched Kosmo, a 3D world data engine consisting of the handheld AI spatial camera Kosmo One and the Kosmo World cloud platform. The former captures real environments, while the latter converts the captured results into editable, reusable 3D spatial data.
Huaxia Robotics brought its bionic robot Xialan R03, the IP series Xingxingxia P2, the general-purpose robot Xiaqi D02, and the "RoboEase™" scenario platform to the 2026 Physical AI and World Models Conference. Xiaqi D02 completed recognition, picking, delivery, and placement at designated points after self-service ordering at an intelligent beverage shelf, with no manual intervention throughout; Xialan R03 handled multilingual welcome reception and guided explanations at the outdoor venue. The RoboEase™ cockpit interface supports monitoring of robot positions and status across the venue as well as multi-robot task scheduling.
Afferent-Omni is an open-source visuo-tactile command and control (C2) framework for embodied intelligence and dexterous manipulation, offering real-time visuo-tactile perception, closed-loop manipulation, and embodied AI evaluation capabilities. Its FastAPI backend includes 9 modular routers and 15 REST endpoints, and achieves closed-loop operation via a 50 Hz WebSocket stream; it integrates a neuromorphic reflex arc with under 15 ms latency, multi-finger topology, and acoustic friction profiles.
Dream Labs announced it has completed a $90 million seed round led by B Capital, with participation from Kindred Ventures and others. The company also launched Dream API, which can convert robot camera footage and natural-language instructions directly into motor commands, and said its flagship model Dream-1 is trained on videos of everyday human tasks and can zero-shot control any robot with a gripper.
Why it matters: Dream-1's cross-hardware zero-shot control and the open approach of Dream API provide a concrete reference for judging the transferability of general-purpose robot models.
DeepMirror released a full-stack physical AI solution for the steel industry at the 2026 Unitree Technology Partner Ecosystem Conference held in Hangzhou on October 22. Using Unitree quadruped robot dogs as the carrier and paired with its self-developed Insight series spatial intelligence cameras, the Navcore spatial intelligence brain, and a task management platform, the solution enables a complete inspection chain covering pure vision positioning, autonomous navigation, dynamic obstacle avoidance, equipment detection, and data archiving. The solution has been deployed at multiple steel enterprises including Xiangtan Iron and Steel and Jinnan Iron and Steel, has entered a phase of regular on-site operation, and is planned for reuse in scenarios such as power cable tunnels and chemical industrial parks.
The authors propose LENS, which accelerates ACEsplat-style scene memory workloads on an AMD KR260 FPGA/ARM platform, comprising an INT16 scene coordinate regression accelerator and a fixed-point ordered 3DGS renderer.
The authors propose VERID, a persistent, high-fidelity semantic digital twin for automated laboratories that provides AI agents with four core services: queryable introspection, time-compressed simulation, comprehensive trajectory persistence, and a safe execution framework. Three case studies using mixtures of water-soluble dye and deionized water validated its optical and fluidic foundations, demonstrating pre-detection of critical protocol errors, in-run stateful error recovery, and post-hoc identification of subtle physical faults.
The paper proposes a natural-language policy interface that connects operator instructions to ROS 2 Nav2: a local large language model converts instructions into JSON policies specifying destinations, spatial constraints, and temporal or event conditions; deterministic checks then decide whether to accept them, and three costmap adapters inject them into the existing navigation stack without changing the planner or controller.
RAISE-Vision is an automated platform for contact angle (CA) measurement on curved hydrogel surfaces, intended for in vitro wettability analysis of contact lenses. Its 3D-printed lens mounts, camera holders, and drying apparatus enable an automated, repeated sessile-drop workflow within a liquid handling system. A custom open-source image processing script combines elliptical fitting and the tangent method to handle droplet deformation on curved surfaces. In 10 consecutive measurements of a senofilcon A lens in air, the contact angle rose from 62.6±5.5° to 83.4±2.1°, with no significant difference from the manual protocol results (p=0.470).
This paper investigates the coupling between numerical solver depth and execution horizon in flow-matching vision-language-action (VLA) models, finding that coarse-grained numerical integration reduces accuracy under rapid replanning but instead improves task performance under extended open-loop execution.
The paper points out that when VLA models are fully fine-tuned in bf16 without keeping an fp32 master copy, AdamW rounds away any update smaller than 1/512 of a weight; 9 of 12 open-source VLA codebases train this way by default or in documented recipes, including LeRobot's SmolVLA default configuration and the π0/π0.5 guides.
Why it matters: The original quantifies the proportion of updates lost in bf16 fine-tuning and provides a LIBERO success-rate comparison for the fp32 master copy, which can serve as a reference for training configurations.
Space intelligent robotics company Jinkong Technology has completed an angel round of tens of millions of yuan, led by CDH VGC with participation from Shanghai Jiao Tong University's Hanyuan Asset, Guoke Green Fund, and Lingqiao Intelligent. The funds will be used for R&D of core components for space robots, in-orbit verification, and product engineering.
On October 23, Anqing Liandong Attachments, a subsidiary of Anhui Forklift Group, released a matrix of eight categories of electric-drive smart attachment products in Anqing, replacing hydraulic structures with electric actuators and covering mainstream industrial scenarios such as electric three-way forks, electric fork positioners, and electric rotators. The series is controlled by electrical signals, can connect with AGV/AMR and intelligent logistics scheduling systems, is compatible with unmanned handling equipment from multiple brands, and eliminates the risk of hydraulic oil leaks while reducing oil-line maintenance. After the launch event, customers observed working-condition simulations on site and toured the production workshop.
Without modifying the AGV PLC program, the Kollmorgen NDC scheduling system uses configurations such as TS, Cluster, Blocking, Weight, and SSIO in System manager/System manager NG to implement three traffic management solutions for oncoming traffic on narrow paths.
A developer built Corebrum OS for physical AI over a weekend on an Ubuntu Server 26.04 base, integrating the Hyprland desktop, ROS 2 with AgenticROS, RealSense 3D perception, and the Zenoh middleware.
Heli Yufeng has designated semiconductors as a strategic focus industry and, based on "stereoscopic storage + AGV handling and delivery," is building a full-chain logistics coordination system covering raw material storage, production line delivery, material picking, finished goods storage, and intelligent truck loading. The company has delivered an intelligent warehousing and conveying system integrating automatic depalletizing, inbound storage, picking, and outbound storage to a semiconductor chip manufacturing enterprise, and has provided an automatic packaging + intelligent warehousing solution for a semiconductor materials enterprise, achieving automated linkage between packaging and warehousing for anti-static protection and full-batch traceability.
OSWAM proposes a world action model with one-step action generation. By constructing imagination tokens to initialize the diffusion process and providing physical priors and action-conditioned representations, it compresses the tens of denoising steps required by direct-policy world action models into a single step.
A developer set up a robot lab in an office Gaussian splat reconstruction scene, having a borrowed RealMan ECO65 perform a J1 joint +5° round-trip motion at 5% speed; after switching to Wi-Fi client mode, it lost connection and no Ethernet cable was available.
During the 2026 QinCheng Conference, the Amap Momentum robot dog Tutu took part in guided explanations and tours, and together with Alibaba Group General Affairs and QinCheng Smart, released a smart campus case study and joint solution, bringing Tutu's inspection work at the Hangzhou Xixi Campus into the spotlight. Amap divides the ABot system into three layers—DATA, MODEL, and AGENT—supported by the ABot-N1 navigation base and the ABot-M0.5 manipulation base, and adopts a dual-speed architecture of cloud-based macro planning plus on-device embodied execution.
Wouf, the quadruped robot in the open-source Cubic Doggo project, has completed its first walk. The author traced the earlier low-speed problem to latency in reading position, speed, and torque from the ST3215 servos (about 3–5 ms per read across three servos) and has now disabled feedback reading on Wouf. After switching to the ST3215, Wouf's walking speed exceeded expectations, but the turning algorithm still has issues. The author believes the approach can be used for walking robots, and whether feedback reading (an input required for reinforcement learning) can be restored depends on how difficult it is to develop synchronized-read firmware.
AgiBot organized 110 full-size Expedition A3 humanoid robots to dance together at the Yellow River International Convention and Exhibition Center in Jinan, Shandong, setting a Guinness World Record for "most humanoid robots dancing simultaneously." The adjudicator on site confirmed that the successful count was 110 robots.
A user bought an old Beam+ telepresence robot from FB Marketplace; its battery was dead, and after powering it with a desktop power supply it booted successfully, but authentication could not be completed because the software points to a server that has been taken offline. They asked Awabot about a software update or system image, and also found papers and GitHub repositories referencing the old ROS project and added sensors, as well as a video showing commands sent manually to drive the robot and capture its video stream.
A developer is building a robot consisting of a Raspberry Pi 5, an ESP32, and 2 RS390 6V motors, powered by a 3S4P pack of 12 18650 cells (2,200mAh each), a 40A BMS, nominal 11.1V (12.6V fully charged). The core question is how to provide the Pi 5 with a 5.1V supply that has sufficient current under load: whether to choose a high-current step-down converter or whether there is a better solution, and the developer hopes for specific module recommendations.
HOJ Innovations has deployed a fleet of autonomous mobile robots (AMRs) at Hunter Douglas's Salt Lake City fulfillment center, replacing the existing overhead conveyor system (OCS).
STIAL has built four product series around complex grinding scenarios: AI flexible grinding robots, AI flexible grinding workstations, embodied core components, and the NextBrain™ AI brain, forming a physical AI technology loop of "perception—analysis—decision—execution."
NVIDIA released an illustrated tutorial that uses a fictional startup building a warehouse humanoid robot as an example, connecting tools including Isaac Sim, Isaac Lab, GR00T, Cosmos, and Omniverse, and covering virtual environment setup, demonstration teaching, policy training and testing, handling the Sim2Real gap, and deploying AI to real hardware. The tutorial also explains the steps between an AI model deciding to move an arm and the motor actually executing it, and the author hopes to exchange with developers who have actually used these tools about how real workflows differ from the examples.